Pathogen and Indicator Organisms Removal in Artificial Greywater Subjected to Aerobic Treatment

– Comparison of four filter media

Ramiyeh Molaei

Institutionen för energi och teknik Department of Energy and Technology

Examensarbete 2014:03 ISSN 1654-9392

Uppsala 2014

English title: Pathogen and indicator organisms removal in artificial greywater subjected to aerobic treatment

Swedish tittle: Reduktion av patogener och indikatororganismer i aeroba BDT-vattenfilter - Jämförelse mellan fyra filtermedia

Ramiyeh Molaei

Supervisor: Cecilia Lalander, Swedish university of agriculture,Energy and technology

Björn Vinnerås, Swedish university of agriculture,Energy and technology

Examiner:

Credits: 30 hecLevel: Second cycle, A2ECourse title: Independent project in environmental scienceCourse code: EX0431Programme/education: soil and water management

Place of publication: UppsalaYear of publication: 2014Cover picture: Ramiyeh MolaeiOnline publication: http://stud.epsilon.slu.se

Keywords: activated carbon, pine bark, biochar, BOD5 reduction, Enterococcous faecalis,greywater reuse, bacteriophage MS2, bacteriophage ɸX174, Salmonella spp.

Sveriges lantbruksuniversitet

Swedish University of Agricultural Sciences

Faculty of Natural Resources and Agricultural Sciences

Energy and technology

Abstract

Treated greywater has a good potential to be used as a source of water. This

study analyzes the removal of Salmonella spp., Enterococcous faecalis,

bacteriophage ɸX174 and bacteriophage MS2 from greywater by filtration

through biochar, bark, activated carbon and mixture of bark and activated

carbon. Reduction of pathogen and indicator organisms were studied over a

period of 63 days in column experiments (height 65 cm, diameter: 4.3 cm)

supplying artificial greywater at a hydraulic loading rate of 0.032 m3 m

-2 day

-1

and an organic loading rate of 76 g BOD5 m-2

day-1

(240 g COD m-2

day-1

).

Biochar filters were more effective than other filters in removal of

Salmonella spp. (3 log10 reduction) and were less effective in removal of

bacteriophages. Bark and mixture filters performed inefficiently in removal of

pathogen and indicator organisms (1-2 log10 reduction). High reduction of

pathogen and indicator organisms was detected in activated carbon filters

within the first half of the experimental period. The reduction was 7 log10 for

Salmonella spp., 5 log10 for E. faecalis, 6 log10 for bacteriophage фX174 and

3 log10 for MS2. The reduction of Salmonella spp. and ɸX174 correlated to the

inflow concentration in all filter media.

Keywords: activated carbon, pine bark, biochar, BOD5 reduction, Enterococcous

faecalis, greywater reuse, bacteriophage MS2, bacteriophage ɸX174, Salmonella spp.

Author’s address: Ramiyeh Molaei, SLU, Department of Energy and Technology,

P.O. Box 7032, 75007 Uppsala, Sweden

E-mail: mora0004@stud.slu.se

Foreword This master thesis has been conducted at the Department of Energy and

Technology at the Swedish University of Agricultural Science in Uppsala,

Sweden as a part of a research project that focused on on-site greywater

treatment in order to reuse for irrigation, service or recharge of surface and

groundwater. It was realized by funding from the Swedish International

Development Cooperation Agency (Sida) and the Swedish Research Council

(Formas).

I would like to sincerely thank my supervisor, Cecilia Lalander (Department of

Energy and Technology/SLU), for her cheerful continuous supervision. I wish

to thank Björn Vinnerås (Department of Energy and Technology/SLU) for

valuable advices and support. I also want to thank Sahar Dalahmeh

(Department of Energy and Technology/SLU), Annika Nordin (Department of

Energy and Technology/SLU) and Jörgen Fidjeland (Department of Energy

and Technology/SLU) for offering their professional laboratory experiences.

At last, I want to thank Yury Chaiko for his help and encouragement.

Filthy water cannot be washed.

West African Proverb

Contents Abbreviations 9

1 Introduction 11 1.1 Objective 12 1.2 Hypothesis 12

2 Background 13 2.1 Greywater characteristic and analysis 13

2.1.1 Physical characteristics 13 2.1.2 Chemical characteristics 14 2.1.3 Pathogenic characteristics 15

2.2 Greywater treatment 17 2.3 Greywater reuse guidelines 18

3 Material and Methods 20 3.1 Experimental design 20 3.2 Filter materials 20 3.3 Artificial greywater 23 3.4 Pathogen and indicator organims 23 3.5 Pathogenic and indicator organisms analysis 24 3.6 Statistics 24

4 Results 25 4.1 Residence time and tracer recovery 25 4.2 pH values 27 4.3 Reduction of pathogen and indicator organims 28

4.3.1 The pathogenic and indicator organisms concentration in the inflow 28

4.3.2 Salmonella spp. 29 4.3.3 Enterococcous faecalis 29 4.3.4 Bacteriophage фX174 30 4.3.5 Bacteriophage MS2 31

4.4 Log reduction and inflow concentration 32 4.4.1 Salmonella spp. 32 4.4.2 Enterococcous faecalis 33 4.4.3 Bacteriophage фX174 33 4.4.4 Bacteriophage MS2 34

4.5 Statistic 35 4.6 Pathogen and indicator organisms residence time 37

4.6.1 Salmonella spp. 37 4.6.2 Enterococcous faecalis 37 4.6.3 Bacteriophage фX174 38 4.6.4 Bacteriophage MS2 39

5 Discussion 41 5.1 Organic loading of greywater 41 5.2 Filter material 41 5.3 Residence time and tracer recovery 42 5.4 Pathogen and indicator organisms reduction 43

5.4.1 Bacterial reduction 45 5.4.2 Bacteriophage reduction 46

5.5 Log reduction and inflow concentration 48 5.6 Pathogen and indicator organisms residence time 48 5.7 Reuse possibilities of filtered water 49

6 Conclusion 50

References 51

9

Abbreviations AC Activated carbon

BC Biochar

B Bark

BOD5 Five day biochemical oxygen demand

COD Chemical oxygen demand

CFU Colony forming unit

EC Electrical conductivity

E. coli Escherichia coli

E. faecalis Enterococcous faecalis

ISP Isoelectric pH

M Mixture of activated carbon and bark

PFU Plaque forming unit

10

11

1 Introduction Water is a finite precious resource (UNU-INWEH, 2011-2012, Anwar, 2011)

which is essential for multidimensional development (FAO, 2007).

Low rainfall (Eriksson et al., 2002), population growth, increased urban,

agricultural and industrial water demand (UNU-INWEH, 2011-2012, Rose et

al., 1991) in combination with unbalanced water demand and availability,

accentuate the water issue (FAO, 2007). Water stress is a serious challenge for

many countries all over the world (Li et al., 2010, Pinto et al., 2010) specially

in arid and semi-arid areas (Santos et al., 2012, FAO, 2010). Around one third

of the world’s population face water shortage and scarcity, but this number is

estimated to increase to two-third by 2025 (FAO, 2007).

The highest demand for water, 70%, come from agriculture and it is estimated

to double by 2050 (FAO, 2007, FAOWATER, 2008). Consequently the wise

use of water is a key for sustainable development (FAO, 2007).

The consumption of fresh water can be decreased between 30% to 50% (Pinto

et al., 2010, Jeppesen, 1996) by using greywater for toilet flushing, cleaning

purposes, industrial cooling processes and agricultural irrigation (Sellner, 2009,

Pinto et al., 2010). Greywater is the wastewater produced in kitchen sinks,

bathtubs, showers, hand basins and laundry machines (Eriksson et al., 2002).

Despite the fact that faeces, urine and toilet papers are absent in greywater, it

requires pre-treatment before irrigation, for health and environmental reasons

(Eriksson et al., 2002). Irrigation with untreated greywater can lead to spread

of disease (Eriksson et al., 2002), dissolution of organic matter (Anwar, 2011),

reduced yield (Chen et al., 2000, Morel and Diener, 2006), oxygen depletion

(Eriksson et al., 2002), eutrophication (Morel and Diener, 2006) and separation

of soil particles (Anwar, 2011).

12

1.1 Objective

The objective of this study was to compare the efficiency of four filter

materials: biochar, bark, activated carbon and mixture of bark and activated

carbon, in removal of pathogen and indicator organisms, Salmonella spp.,

Enterococcous faecalis, bacteriophage ɸX174 and bacteriophage MS2, from

artificial greywater.

1.2 Hypothesis

Concentration of pathogens and indicator organisms in artificial greywater can

be reduced through vertical biofilter columns.

Biochar filters can provide high removal of pathogens and indicator organisms

due to their big effective size similar to bark, high retention time, surface area

and porosity. Biochar material might be a practical alternative for activated

carbon material due to its simple and low cost production comparing to

activated carbon.

Bark filters can significantly remove the pathogens and indicator organisms as

previous studies illustrated effective reduction in bark filters due to their

chemical composition (rich in tannins, low pH) and surface charge.

Activated carbon particles have high specific surface area, porosity and

specific surface which lead to high biofilm formation potential and adsorption

capacity. Hence, it included in this experiment as a reference.

Mixture of bark and activated carbon can perform significant reduction of

pathogens and indicator organisms since both filter materials are effective and

also can decrease the installation cost due to reducing the use of expensive

activated carbon.

13

2 Background The use of greywater in agriculture has become an interesting option in order

to reduce the use of freshwater, but it requires treatment in order to insure the

human and environmental safety (FAO, 2010).

2.1 Greywater characteristic and analysis

The share of greywater in total volume of wastewater , in which toilet water is

included, is around 50-80% (Al-Jayyousi, 2003, Li et al., 2010). The

concentration of organic matter, nitrogen, phosphorous, salts, solids and

pathogens is extensively different among households depending on type of the

detergents, type of goods being washed, food diet and life style of residents

(Pinto et al., 2010, Eriksson et al., 2002). The parameters that should be

analyzed before irrigating greywater on land are pH, electric conductivity

(EC), suspended solids, heavy metals, dissolved oxygen, biological and

chemical oxygen demands (BOD and COD), pathogens and indicator

organisms ; fecal coliforms, E.coli; total nitrogen and phosphorus, (Pinto et al.,

2010, Negahban-Azar et al., 2012).

2.1.1 Physical characteristics

Suspended solids in high concentration can clog the filters and irrigation

suppliers (Eriksson et al., 2002). Sources of solids in greywater are food

material, hair and fibers (Eriksson et al., 2002). Eriksson et al. (2002) reported

the solids concentration to be between 113 and 2410 mg L-1

.

14

2.1.2 Chemical characteristics

Alkalinity and pH

The alkalinity of the greywater is usually in a range of

20-340 mg CaCO3 L-1

and is highly related to the pH of water supply. The pH

in greywter originated from laundry is higher than other fractions of greywater,

around 8-10, due to the presence of chemical products (Morel and Diener,

2006). Buffering capacity and pH of the soil can be effected by the alkalinity

and pH of the greywater (Eriksson et al., 2002). High pH (>10) can cause

dissolution of organic matter, plant suffer and soil particles separation (Anwar,

2011).

Organic load

The main organic substances found in greywater are proteins, carbohydrates,

fats and surfactants (Morel and Diener, 2006). Organic load is measured by the

chemical oxygen demand (COD) and/or biological oxygen demand (BOD) and

vary between 13 and 8000 COD mg-1

L-1

or between 5 and

1460 BOD mg-1

L-1

. COD mainly originates from dishwashing and laundry

detergents (Eriksson et al., 2002). The source of oil is kitchen and is related to

the cooking habits. The BOD and COD shows the organic matter load in the

water which in high load can lead to oxygen depletion and sulphide production

due to activity of anaerobic bacteria (Eriksson et al., 2002). The fat can congeal

and stop the irrigation or treatment process (Morel and Diener, 2006). It shows

the particle content which can lead to clogging the filters used for greywater

treatment or irrigation suppliers (Eriksson et al., 2002).

The main source of nitrogen in greywater is kitchen waste, 40-74 mg L-1

.

Bathroom and laundry have the minimum share of nitrogen in greywater. The

source of phosphates in greywater is mainly washing detergents, in the

countries that are still using phosphorus-containing detergents, and the total

phosphorus concentration is around 6-23 mg L-1

. In countries that have banned

phosphorus-containing detergents, the concentration is about 4-14 mg L-1

,

which mainly originate from bathroom greywater (Eriksson et al., 2002).

Despite the fact that nitrogen and phosphorus are good fertilizers they have

negative effects on aquatic environment. If spread into water bodies they can

cause algae growth and eutrophication (Morel and Diener, 2006).

15

Inorganic load

Inorganic components- Calcium carbonate (CaCO3) and some other inorganic

salts such as sodium chloride (NaCl) are present in greywater (Eriksson et al.,

2002). The electric conductivity of greywater which shows the salinity of the

greywater is between 300-1500 µS cm-1

and sometimes up to 2700. Irrigation

with saline greywater in the first place can decrease the yield and in the long

term can lead to topsoil salinization in arid regions with clay and loamy soils

(Morel and Diener, 2006). The EC of irrigation water is usually about 1 ds m-1

(Anwar, 2011) and recommended range is 0-1500 µS cm-1

(Pinto et al., 2010).

Essential mineral elements- Manganese (Mn), iron (Fe), copper (Cu), nickel

(Ni), zinc (Zn) can be found in greywater, for instance concentration of zinc is

around 0.01-1.8 mg L-1

or less (Eriksson et al., 2002). Despite the living

organism requirement to such elements, excessive levels can have damaging

effects on them.

Non-essential mineral elements- Aluminum (Al) and heavy metals; cadmium

(Cd), lead (Pb), mercury (Hg), chromium (Cr) can also be found in low

concentrations in greywater (Eriksson et al., 2002). Human exposure to such

elements can cause physical and mental problems. The high concentration of

heavy metals can affect the quality and quantity of the yield as well as aquatic

environment (Chen et al., 2000) and the risk of metal transportation to the

groundwater is related to physical, chemical and biochemical properties of the

soil (EPA, 2002). Accumulation of heavy metals in animals can cause serious

illness in long term (EU, 2009).

The source of mineral elements (except P and N) can be plumbing materials,

jewelry, cutlery, coins and home maintenance products (Eriksson and Donner,

2009).

2.1.3 Pathogenic characteristics

Greywater is less contaminated in term of pathogens compared to wastewater

since the toilet water is excluded. Although it is available in higher volume

with lower level of microbial load (Pinto et al., 2010). Pathogens such as

bacteria, viruses, protozoa, parasites can be found in grewywater (Morel and

Diener, 2006) . Several studies observed fecal contamination in greywater

(Jeppesen, 1996, Ottoson and Stenström, 2003, O'Toole et al., 2012, Rose et

al., 1991) which raise health concerns on irrigation with greywater (Negahban-

16

Azar et al., 2012). Fecal coliforms concentration in greywater varies from

0 to 106-10

7 CFU 100 mL

-1 (Friedler et al., 2006).

Pathogenic viruses and bacteria can reach the greywater in different ways.

Greywater originated from kitchen can be contaminated by uncooked

vegetables and raw meat and it contains higher levels of micro-organism

comparing to laundry and bathroom greywater. Escherichia coli concentration

in the greywater was found in the range of 2.5×105-1.3×10

8100 mL

-1 arisen

from kitchen. The laundry greywater contains total coliforms in range of

8.9×105-56 ×10

5100 mL

-1. The bathroom greywater contains total coliforms in

range of 70-2.4×107100mL

-1 which can be introduced to greywater by hand

washing after toilet visit, diaper changing, bathing children (Eriksson et al.,

2002) and anal cleansing.

In this study Salmonella spp. were used as fecal pathogen model and indicator

organisms Enterococcous faecalis and bacteriophages ɸX174 and MS2 were

used as model organisms.

Bacteria divided into two main groups, gram-negative and gram-positive,

based on the presence or absence of an outer lipid membrane. Outer lipid

membrane is present in gram-negative bacteria which make the wall

impenetrable and thus more resistant against antibodies (Murray et al., 2002).

Salmonella genus belongs to the family Enterobacteriaceae and comprises

over 2600 serovars, which are divided into typhoidal and non typhoidal

serovars. Typhoidal Salmonella cause systemic invasive disease (enteric fever)

by the fecal-oral route and via contaminated food and water. Non typhoidal

serovars cause less sever, but commonly occurring gastroenteritis

(salmonellosis), mainly in animal production (poultry and eggs). Salmonellosis

infection circle begins with the bacteria entering and localizing the host

intestines within 12 to 72 hours and results in diarrhea, nausea, vomiting and

fever (Rhen, 20007). Salmonella spp. are gram-negative, motile, rode-shaped

(0.7-1.5 × 2-5 µm), facultative anaerobic, zoonotic intestinal (Rose et al., 1991)

with 2-5 µm size.

Enterococcous faecalis genus is part of Enterococci family, gram-positive,

spherical or ovoid shaped (0.6-2.0 × 0.6-2.5 µm), occurring in pairs or short

chains, facultative anaerobe, non-spore forming with 0.5-1 µm size (Bergey

and Holt, 1994). E. faecalis live in gastro intestinal of humans and other

mammals (R.Eley, 1996) and its presence in water bodies and foods inferred to

17

fecal contamination (Riemann and Cliver, 2006) . This, together with tolerance

to extreme conditions such as hyperosmolarity, heat (growing at 10°C and

45°C, surviving at 60°C for 30 minutes), ethanol, hydrogen peroxide, acidity,

alkalinity (growing at pH 9.6) and salinity (6.5% NaCl broth) (Güven

Kayaoglu and Ørstavik, 2004) makes E. faecalis good indicators.

Viruses pose serious health risk due to their relatively low infectious dose

(Dixon et al., 1999). Constant presence of bacteriophages in treated sewage

(Henze, 2008) in addition to theirs inexpensive, simple and quick analyze and

being non-pathogenic to humans propose them as good fecal and viral

indicators (Lalander et al., 2012, Ottoson, 2004). Bacteriophages have been

wildly used to examine the effect of water and wastewater treatments on

viruses (Henze, 2008).

Bacteriophage ɸX174 is an icosahedral shaped somatic coliphage (attach to the

host cell wall), with 27 nm diameter and isoelectric pH between 6.6 and 6.8.

Previous studies showed high rate of adsorption to filter media which makes it

a reliable model organism for viruses (Collins et al., 2006).

Bacteriophage MS2 is an icosahedral shaped f-specific bacteriophage (infect

the host cell appendage sex pili), 27 nm in size and with an isoelectric pH

around 3.5. Even though previous studies showed low levels of MS2

adsorption to filter media it is considered as a convenient model organisms for

viruses (Collins et al., 2006).

2.2 Greywater treatment

Greywater is considered a reliable source of irrigation in The World Health

Organization Guidelines for safe use of wastewater, excreta and greywater

(WHO, 2006a) due to the considerable level of plant nutrients and low

concentration of pathogens comparing to wastewater. Common method for

treating greywater is filtering system; various filter media have been explored.

Sand filtering is the most common approach for treating the greywater

(Dalahmeh et al., 2012). Despite the fact that sand filters can reduce the BOD5

by 75% in vertical sand filters, investigations into alternative filter materials

have started due to the clogging and accessing problems of sand in some

regions, considering that the high bulk density of sand make it costly to

transport (Dalahmeh et al., 2012).

Horizontal subsurface flow constructed wetland (using sand or oil-shale ash as

a filter media) is another method which is reliable to remove P and N from

18

wastewater but efficiency of this system decreasing by time due to saturation,

clogging and biofilm formation (Vohla et al., 2007).

In 1990s vertical flow systems became very popular in Europe (Stefanakis and

Tsihrintzis, 2012) and several studied examine their efficiency in treating the

greywater. Coetzee et al. (2010) studied aerobic treatment using PVC vertical

columns (150 cm×15 cm) filled with stone. The best result shown in term of

nitrogen removal from greywater at hydraulic load 35.7 L m-2

d-1

(Coetzee et

al., 2010) whereas it was more effective in removal of nitrogen and phosphorus

in higher organic load (up to 200 g COD m-2

). Zapater et al. (2011) showed

that re-circulating vertical flow constructed wetland (RVFCW) is reliable in

term of nitrogen, phosphorus, BOD and COD excepting E. coli. In another

study by Stefanakis and Tsihrintzis (2012) carbonate and igneous rock, zeolite

and bauxite used as filter material in vertical flow constructed wetlands and the

results showed that this system is efficient in BOD5 ,COD (up to 78%) and

nitrogen removal but not efficient in removal of phosphorous.

Dalahmeh et al. (2012) compared both pine bark and activated carbon with

sand using the vertical PVC columns. In this study, bark and charcoal filters

performed higher BOD5, P and N removal and fecal coliform reduction than

sand filters. Lalander et al. (2012) study and Lewis et al. (1995) showed that

bark material can efficiently remove the pathogens from greywater.

Nonetheless the efficiency of bark, charcoal and similar carrier materials

requires further investigation.

2.3 Greywater reuse guidelines

According to WHO guidelines for reuse of greywater in agriculture, acceptable

values of E. coli and thermotolerant coliforms in irrigation water are

103-10

5 100 mL

-1 and 10

4-10

6 100 mL

-1 respectively. WHO suggests different

combinations of greywater treatment, crops and irrigation system in order to

achieve a total pathogen reduction of 7 log10. The required level of reduction

by treatment varies between 1 to 4 log10 reduction (WHO, 2006b).

In restricted irrigation which 7 log10 reduction can be achieved in a number of

scenarios: (I) treatment (4 log10 reduction) and labor intensive

(3 log10 reduction); (II) treatment (3 log10 reduction) and highly mechanized

system (4 log10 reduction); (III) treatment (1 log10 reduction) and subsurface

irrigation (6 log10 reduction); (IV) cooking (6-7 log10 reduction). Peeling before

consumption can provide 2 log10 reduction. Restricted irrigation includes non-

19

food crops (e.g. cotton and biodiesel crops), food crops that are processed

before consumption (wheat) and all crops that have to be cooked before human

consumption (potatoes, rice) (WHO, 2006b).

In unrestricted irrigation, which includes crops that are eaten uncooked by

humans, leaf crops (lettuce) and root crops (onion) require a total reduction of

(6 and 7 log10 reduction), which can be achieve by various scenarios:

(I) treatment (3 and 4 log10 reduction), pathogen die-off on crop surfaces

(2 log10 reduction) and washing of with clean water prior to consumption

(1 log10 reduction). High growing crops and low growing crops require

6 log10 reduction, which can be achieved by: (II) treatment (2 log10) and drip

irrigation (4 log10) for high crops (such as tomato) and (III) treatment (4 log10)

and drip irrigation (2 log10) for low crops (WHO, 2006b).

Among different irrigation systems, flood and furrow irrigation are the lowest

cost methods but highest risk to the fieldworkers. Spray and sprinkler method

have highest potential to spread the contamination on to crops and field

workers (just 1 log10 reduction). Drip irrigation considered as the most health

secure method for fieldworkers and the most expensive one. However,

reducing the greywater consumption and increasing the crop yield may recover

the expenses (WHO, 2006b).

20

3 Material and Methods

3.1 Experimental design

The experiment took place over a period of 78 days. For filter materials

prepared and a total of twelve vertical columns were installed with triplicates

of each filter material. The bulk density, particle density, total porosity,

gravimetric water content, residence time and tracer recovery of each filter

material was determined before feeding the columns with greywater. The

columns were fed with artificial greywater over a period of 63 days and

duplicate microbial analysis performed for each column.

3.2 Filter materials

Four filter materials; biochar, bark, activated carbon and mixture of bark and

activated carbon were sieved to meet particle size distribution correlated to

previous studies by Dalahmeh et al. (2012) and Berger (2012) .

The biochar originated from salix wood in Germany were sieved through 7, 5,

2.8, 1.4 and 1 mm screens. Biochar retained on the 2.8 and 1 mm screens was

mixed a ratio 3:2 by weight. The bark originated from an undefined air-dried

pine bark and sieved through 7, 5, 3.15 and 1 mm screens. Bark retained on the

3.15 and 1 mm screens was mixed in a ratio 3:2 by weight. The activated

carbon obtained from Merck with two different sizes; 1.5 and 3-5 mm length

and 1.5 mm diameter. The material was mixed in a ratio 2:3 by weight. The

mixture material was prepared by mixing the bark and activated carbon in a

ratio 1:1 by weight.

21

The vertical columns that were used in the experiment were made of

transparent acrylic plastic with a height of 65 cm, diameter of 4.3 cm and

bottom outlet with 0.5 cm diameter. The columns were covered by aluminum

foil in order to inhibit algae growth. Perforated aluminum foil was used as lid

for the columns.

A total of twelve columns were installed and filled up to a depth of 50 cm with

biochar, bark, activated carbon and mixture, with three columns for each

material. The filter material was added spoon by spoon and shaken manually in

order to have higher density (Figure1). Two layers of 10-25 mm gravel were

placed under and over the filter material in order to facilitate drainage

(Figure1).

Figure 1.Installation of experimental column with filter material, top and bottom

gravel and greywater application.

22

The bulk density, particle density, total porosity, gravimetric water content and

residence time of the filter material were determined after the column

installation (Table 1).

Bulk density was calculated by dividing the dry weight of the filter material by

the total volume of cylinder occupied by the material (50 cm depth and 4.3 cm

diameter).

Particle density of filter material was determined by dividing the dry weight of

material by the volume of solids excluding the pores.

The liquid immersion method was used in order to measure the volume of

deionized water displaced by particles. The mixture of dry material and

deionized water was boiled gently in a volumetric flask in order to determine

the air-filled pores and cover flask remained in lab for 24 hours to saturate and

they filled up with water. The weight of the flask was recorded in all the steps.

Porosity was calculated by:

where ρB is the bulk density and ρp is the particle density.

Gravimetric water content was determined by dividing the mass of water by

the weight of oven dried material. The mass of water was calculated by

subtracting the weight of the oven dried material from the weight of the air

dried material.

Residence time was determined after saturating the filter material by

immersing the columns in tap water. The process started by adding a 0.03 L

pulse of NaCl tracer solution (10 g L-1

) to the columns. Over the period of 11

days columns were fed with 45 mL of tap water (30 mL at 9:00 a.m. and

15 mL at 16:00 p.m.) and the electrical conductivity (EC) of the effluent was

measured an hour after each feeding incident. Mean and longest residence

times of tracer were determined. The mean residence time is the time required

to recover 50% of the tracer and the longest residence time is the time when no

tracer is recovered.

23

Table 1.Characteristics of the filter materials

Filter

material

Particle size

(mm)

Effective

size

(mm)

Bulk

density

(g m-3

)

Particle

density

(g m-3

)

Total

porosity

(%)

Water

content

(%)

Biochar 1-1.4 2.8-5

1.4 0.27 0.74 63.3 0.06

Bark 1-3.15 3.15-5

1.4 0.36 1.3 73 7.6

Activated carbon

1.5 3-5

1.2 0.56 1.89 70.6 1.9

Mixture 1-1.4, 2.8-5 1-3.15, 3.15-5

1.5 0.15 1.42 99.9 6.8

3.3 Artificial greywater

Artificial greywater was prepared by mixing 1.25 g standard nutrient broth

(OXIOD, Sollentuna, Sweden), 0.16 g washing powder (Ariel, Germany),

0.16 g dishwashing gel (YES Original, Procter and Gamble, Stockholm,

Sweden), 0.16 g hair shampoo (VO5, Uplands Väsby, Sweden) and 0.1 g corn

oil (El Nada, Al-Asher for products, 10th Ramadan City, Egypt) in 0.25 L tap

water. The total volume of 3 L greywater was prepared every week and kept in

5 during the week. The temperature of greywater that columns were fed with

was 25 .

3.4 Pathogen and indicator organims

Salmonella enterica subspecies~1 serovar Typhimurium phage type~178

(isolated from sewage sludge) (Sahlstrom et al., 2004) and Enterococcous

faecalis (ATCC 29212) were grown in nutrient broth (Oxiod AB, Sweden) at

37 over 24 and 48 hours. Bacteriophages ɸX174 and MS2 were used as

model organisms. Inoculate solution had 4% volumetric share of the greywater

and was added to 25 greywater every morning.

24

3.5 Pathogenic and indicator organisms analysis

Every week a duplicate analysis was performed for inflow and effluent.

Salmonella spp. were grown on Xylose lysine deoxycholate (XLD) plates and

incubated at 37 for 24 hours. E. faecalis were grown on Slanetz &

Bartley agar (SlaBa) and incubated at 44 °C for 48 hours. Analysis of ɸX174

was performed using E. coli (ATCC 13706) as host bacteria while Salmonella

Typhimurium (WG49, ATTC 700730) host bacteria were used for MS2. The

analysis was conducted by double layer agar method using blood agar base

(BAB) plates as a first layer and soft agar as a second layer. All BAB plates

were incubated at 37 over night. Triple sugar iron-agar method used to

validate typical Salmonella spp. colonies.

3.6 Statistics

The One-way ANOVA test and a post-hoc Tukey’s test were used to analyze

the data using the statistical software Minitab version 10.

One-way ANOVA can show whether there are statistically significant

differences somewhere in the data but cannot show just where those

differences lie. Multiple comparisons test like Tukey’s test in order to point out

just where the real differences lie. The null Hypothesis (H0) was that there is no

significant difference between the means and alternative Hypothesis (H1) was

that there is a significant difference between the means. P-value below 0.05

rejects the null Hypothesis, which means there is a significant difference

between filters.

25

4 Results

4.1 Residence time and tracer recovery

The mean residence time for biochar filter was 90 hours and the longest

residence time was 170 hours (Figure 2, Figure 3) (Table 2).

Figure 2. Net EC vs. time for biochar

1,2 and 3after adding a pulse of NaCl

to influent tap water, loaded at a rate of

0.032 m3 m

-2 day

-1.

Figure 3. Percentage recovery of tracer

vs. time for biochar 1, 2 and 3 after

adding a pulse of NaCl to influent tap

water, loaded at a rate of

0.032 m3 m

-2 day

-1.

In case of bark filters, the mean and the longest residence time were 90 and190

hours respectively (Figure 4, Figure 5) (Table 2).

0

1

2

3

4

5

0 50 100 150 200 250

Net E

C (

mS

cm

-1)

Time (hours)

BC1 BC2

BC3 Ref. BC

0

25

50

75

100

0 50 100 150 200 250

Recovere

d tra

cer

(%)

Time (hours)

BC1 BC2 BC3

26

Figure 4. Net EC vs. time for bark1, 2

and 3 after adding a pulse of NaCl to

influent tap water, loaded at a rate of

0.032 m3 m

-2 day

-1.

Figure 5. Percentage recovery of tracer

vs. time for bark1, 2 and 3 after adding

a pulse of NaCl to influent tap water,

loaded at a rate of 0.032 m3 m

-2 day

-1.

The mean and longest residence time for activated carbon were 180 and

250 hours respectively (Figure 6, Figure 7) (Table 2).

Figure 6. Net EC vs. time for Activated

carbon 1, 2 and 3 after adding a pulse

of NaCl to influent tap water, loaded at

a rate of 0.032 m3 m

-2 day

-1.

Figure 7. Percentage recovery of tracer

vs. time for Activated carbon 1, 2 and 3

after adding a pulse of NaCl to influent

tap water, loaded at a rate of

0.032 m3 m

-2 day

-1.

Mixture filters showed no tracer recovery within 11 days after adding the NaCl

pulse (Figure 8, Figure 9) (Table 2).

0

1

2

3

4

5

0 50 100 150 200 250

Net E

C (

mS

cm

-1)

Time (hours)

B1 B2 B3 Ref. B

0

25

50

75

100

0 50 100 150 200 250

Recovere

d tra

cer

(%)

Time (hours) B1 B2 B3

0

1

2

3

4

5

0 50 100 150 200 250

Net E

C (

mS

cm

-1)

Time (hours)

AC1 AC2

AC3 Ref. AC

0

25

50

75

100

0 50 100 150 200 250

Recovere

d tra

cer

(%)

Time (hours)

AC1 AC2 AC3

27

Figure 8. Net EC vs. time for Mixture 1,

2 and 3 after adding a pulse of NaCl to

influent tap water, loaded at a rate of

0.032 m3 m

-2 day

-1.

Figure 9. Percentage recovery of tracer

vs. time for Mixture 1, 2 and 3 after

adding a pulse of NaCl to influent tap

water, loaded at a rate of

0.032 m3 m

-2 day

-1.

In comparison with other filter materials biochar achieved complete tracer

recovery in shortest period of time and together with bark reached the mean

residence time much earlier (half), compared to activated carbon. Activated

carbon was slower than the other filter materials in term of mean residence

time and full recovery (twice in comparison to biochar and 1.5 times in

comparison to bark) (Table 2).

Table 2.The mean and longest residence time for biochar, bark, activated carbon and mixture.

Filter material Mean residence time

(hour)

Longest residence time

(hour)

Biochar 90 170

Bark 90 190

Activated carbon 180 250

Mixture no recovery no recovery

4.2 pH values

In biochar, activated carbon and mixture filters, pH was fairly constant during

the experiment. In biochar filter pH was slightly below 9, in activated carbon

filters was around 9 and in mixture filters was around 8. Bark filters showed

high variation of pH between 4 to up to 7 (Figure 10).

0

1

2

3

4

5

0 50 100 150 200 250

Net E

C (

mS

cm

-1)

Time (hours)

M1 M2 M3 Ref. M

0

25

50

75

100

0 50 100 150 200 250

Recovere

d tra

cer

(%)

Time (hours)

M1 M2 M3

28

Figure 10. pH of biochar, bark, activated carbon and

mixture filters during the experiment.

4.3 Reduction of pathogen and indicator organims

4.3.1 The pathogenic and indicator organisms concentration in the inflow

The concentration of Salmonella spp. and bacteriophages was not constant

during the experimental period and was varying from week to week. However

E. faecalis obtained almost the same concentration over the 60 days of the

operation (Figure 11).

Figure 11. Concentration of pathogens and indicator organims in

the inflow during the experiment. Unit is CFU mL-1

for

Salmonella spp. and E. faecalis and PFU mL-1

for bacteriofaghes.

0

1

2

3

4

5

6

7

8

9

10

0 10 20 30 40 50 60

pH

Time (days)

BC B AC M

1

2

3

4

5

6

7

8

9

10

0 20 40 60

Ave

rag

e lo

g10 c

on

ce

ntr

atio

n

Time (days) Salmonella spp. Enterococcous faecalis

фX174 MS2

29

4.3.2 Salmonella spp.

The reduction in Salmonella spp. was around 2 log10 or higher in the biochar

filter. In activated carbon filter the reduction was very high at the start,

specially during the first week, around 7 log10, but decreased to around 1 log10

after fifteen days. Mixture showed around 1 log10 reduction in the first 30 days

but then increased to 3 log10. Bark filter showed the smallest reduction, 1 log10,

in comparison with three other filter materials over time (Figure 12, Figure 13).

Figure 12. Log10 reduction of

Salmonella spp. in biochar, bark,

activated carbon and mixture at

240 g COD m-2

day-1

.

Figure 13. Percentage reduction of

Salmonella spp. in biochar, bark,

activated carbon and mixture at

240 g COD m-2

day-1

.

4.3.3 Enterococcous faecalis

The effect of biochar filter on E. faecalis reduction was high in the first week,

(4 log10), but decreased to 2 log10 in the second week and less than 1 log10 in

the six following weeks. Bark filter performed low reduction, 1 log10 or less

through entire nine weeks of experience. In activated carbon filter

4 log10 reduction achieved at the start but it decreased to 1 or 2 log10 from day

fifteen. Mixture filters reduction was varying between 0 and 1 log10 (Figure 14,

Figure 15).

0

1

2

3

4

5

6

7

8

0 10 20 30 40 50 60

Log

10 reductio

n

Time (days)

BC B AC M

0

20

40

60

80

100

0 10 20 30 40 50 60

%

reductio

n

Time (days)

BC B AC M

30

Figure 14. Log10 reduction of

E. faecalis. in biochar, bark, activated

carbon and mixture at

240 g COD m-2

day-1

.

Figure 15. Percentage reduction of

E. faecalis. in biochar, bark, activated

carbon and mixture at

240 g COD m-2

day-1

.

4.3.4 Bacteriophage фX174

The removal of the bacteriophage фX174 was the highest in the activated

carbon filter. During the first week it was around 6 log10 but it decreased to

4 log10 for three weeks and 2 log10 in the end. The reduction in bark filter was

around 1 log10 or less for the duration of experiment. Mixture filters performed

similar to bark filters. In the biochar filter the reduction was 4 log10 and

2 log10 in the first and the second week but for the continuing seven weeks of

experiment it was fairly constant, just under 1 log10 (Figure 16, Figure 17).

0

1

2

3

4

5

6

7

8

0 10 20 30 40 50 60

Log

10 reductio

n

Time (days)

BC B AC M

0

20

40

60

80

100

0 10 20 30 40 50 60

% r

eductio

n

Time (days) BC B AC M

31

Figure 16. Log10 reduction of ɸX174 in

biochar, bark, activated carbon and

mixture at 240 g COD m-2

day-1

.

Figure 17. Percentage reduction of

ɸX174 in biochar, bark, activated

carbon and mixture at

240 g COD m-2

day-1

.

4.3.5 Bacteriophage MS2

The reduction of bacteriophage MS2 was similar in both biochar and bark

filter. In the first week the reduction was around 3 log10 in biochar filter and

2 log10 in bark and mixture filters but in all three filters the reduction was less

than 1 log10 during the following nine weeks. Biochar, bark and mixture filters

had highest reduction in the second week. Activated carbon filter had around

3 log10 reduction at the start, but from the day twenty nine it decreased to under

1 log10 (Figure 17, Figure 18).

Figure 17. Log10 reduction of MS2 in

biochar, bark, activated carbon and

mixture at 240 g COD m-2

day-1

.

Figure 18. Percentage reduction of

MS2 in biochar, bark, activated carbon

and mixture at

240 g COD m-2

day-1

.

0

1

2

3

4

5

6

7

8

0 10 20 30 40 50 60

L

og

10 reductio

n

Time (days)

BC B AC M

0

20

40

60

80

100

0 10 20 30 40 50 60

% r

eductio

n

Time (days)

BC B AC M

0

1

2

3

4

5

6

7

8

9

0 10 20 30 40 50 60

Log

10 reductio

n

Time (days)

BC B AC M

0

20

40

60

80

100

0 10 20 30 40 50 60

%

re

du

ctio

n

Time (days)

BC B AC M

32

4.4 Log reduction and inflow concentration

The concentration of four microorganisms in the inflow was not constant

during the experiment. The correlation between the inflow concentration and

reduction of pathogen and indicator organisms, was determined for each

organism by excel regression for the second half of experience since the filters

were more stable from the middle of experiment.

R-square shows how good the regression line approximates the real data and

ideally 0.6 is the least value for R-square. P-value shows the probability of real

results. The lower the P-value, the higher the likelihood that results did not

appear by chance.

4.4.1 Salmonella spp.

Reduction of Salmonella spp. in all filters is highly related to concentration of

Salmonella spp. in the inflow, the higher the concentration the higher the

log10 reduction (Figure 19).

Figure 19. Log10 reduction of Salmonella spp. in biochar, bark, activated

carbon and mixture at 240 g COD m-2

day-1

VS. inflow concentration.

BC

R² = 0.94

P=0.000

B

R² = 0.46

P=0.013

AC

R² = 0.75

P=0.000

M

R² = 0.87

P=0.000

0

1

2

3

4

5

6

7

8

6 6.5 7 7.5 8 8.5 9

Log

10re

ductio

n

Average log10 inflow (CFU mL-1)

BC B

AC M

Linear (BC) Linear (B)

Linear (AC) Linear (M)

33

4.4.2 Enterococcous faecalis

E. faecalis inflow concentration was fairly constant in opposite of the other

microorganisms. The reduction of E. faecalis in relation to inflow

concentration did not show any significant result (Figure 20).

Figure 20. Log10 reduction of E. faecalis in biochar, bark, activated

carbon and mixture at 240g COD m-2

day-1

VS. inflow concentration.

4.4.3 Bacteriophage фX174

Reduction of bacteriophage ɸX174 was highly related to ɸX174 inflow

concentration in all filters (Figure 21).

BC

R² = 0.42

P=0.021

B

R² = 0.22

P=0.119

AC

R² = 0.07

P=0.373

M

R² = 0.20

P=0.138

0

1

2

3

4

5

6

7

8

4.9 5 5.1 5.2 5.3 5.4

Log

10re

ductio

n

Average log10 inflow (CFU/mL) BC BAC MLinear (BC) Linear (B)Linear (AC) Linear (M)

34

Figure 21.Log10 reduction of ɸX174 in biochar, bark, activated carbon

and mixture at 240g COD m-2

day-1

VS. inflow concentration.

4.4.4 Bacteriophage MS2

Reduction of MS2 was not related to MS2 inflow concentration in any of filters

(Figure 22).

BC

R² = 0.88

P=0.000

B

R² = 0.73

P=0.000

AC

R² = 0.58

P=0.025

M

R² = 0.73

P=0.000

0

0.5

1

1.5

2

2.5

3

3.5

3 4 5 6 7

Log

10

reductio

n

Average log10 inflow (PFU/mL)

BC B

AC M

Linear (BC) Linear (B)

Linear (AC) Linear (M)

35

Figure 22. Log10 reduction of MS2 in biochar, bark, activated carbon

and mixture at 240g COD m-2

day-1

VS. inflow concentration.

4.5 Statistic

The result of one-way ANOVA test for 60 days of experiment showed that

there was a significant difference between all filters in reduction of E.faecalis

(P-value=0.009), ɸX174 (P-value=0.003) and MS2 (P-value=0.020), but not in

removal of Salmonella spp. (P-value=0.736). Tukey’s test showed that

activated carbon is significantly different from other filters in E. faecalis and

ɸX174 removal. There was only a significant difference between activated

carbon and mixture in removal of MS2 (Table 3).

Table3. Tuckey’s test for the entire experimental period

Organisms Filter material* Filter material P-value

E. faecalis Activated carbon Biochar 0.0030

Bark 0.0002

Mixture 0.0000

ɸX174 Activated carbon Biochar 0.0000

Bark 0.0000

Mixture 0.0000

MS2 Activated carbon Mixture 0.02

*filter material which is significantly different from the other filter materials

BC

R² = 0.25

P=0.09

B

R² = 0.17

P=0.17

AC

R² = 0.05

P=0.44

M

R² = 0.10

P=0.30

0

1

2

5 6 7 8

Log

10re

ductio

n

Average log10 inflow (PFU mL-1)

BC B

AC M

Linear (BC) Linear (B)

Linear (AC) Linear (M)

36

The concentration of four microorganisms in the inflow was not constant

during the experiment and log10 reduction of organisms was compared at the

highest and lowest inflow concentration during the second half of the

experiment since filters were more stable. Concentration of Salmonella spp. in

the inflow was the highest on the 5th

week and the lowest on 7th

week. Inflow

concentration of E.faecalis was fairly constant comparing to Salmonella spp.

and bacteriophages, however week five and seven were used for analysis.

Concentration of ɸX174 was the highest on week 8 and lowest on week 7,

while the highest concentration of MS2 was observed on week 8 and lowest on

week 6.

The results of one-way ANOVA test showed that there was a significant

difference in reduction of Salmonella spp. (P-value=0.000) (w5, w7),

E.faecalis (P-value=0.000) (w6, w7) and ɸX174 (P-value=0.000) (w7, w8)

between the weeks of highest and lowest inflow concentration for respective

organism, while no difference was established in the case of MS2

(P-value=0.054) (w6, w8). Tukey’s test showed that reduction of Salmonella

spp. was significant at the highest and lowest inflow concentration for all

filters. The same result was established for the reduction of ɸX174. The

removal of E. faecalis was significantly different between week 7 and 8 in all

filters except biochar filters (Table 4).

Table4. Tukey’s test comparing the log10 reduction of each pathogen and indicator organims

between each filter in the highest and lowest inflow concentration

Organisms Comparing weeks Filter material P-value

Salmonella spp. 5 & 7 Biochar *

Bark *

Activated carbon *

Mixture *

E. faecalis 6 & 7 Biochar 0.998

Bark 0.019

Activated carbon *

Mixture 0.002

ɸX174 7 & 8 Biochar *

Bark *

Activated carbon *

Mixture *

*P-value= 0.000

37

At the high inflow concentration of Salmonella spp. (Week 5) biochar

performed similar to bark and mixture filters and different from activated

carbon. Bark filters performance was similar to biochar and activated carbon.

At low inflow concentration of Salmonella spp. (Week 7) all filters performed

the same.

Reduction of E. faecalis at higher inflow concentration (Week 6) was similar in

all filters but at the lower inflow concentration (Week 7) biochar and activated

carbon were different from the other filters while bark and mixture were

similar.

The concentration of ɸX174 was higher in week 8 and the reduction in biochar

and bark filters were similar to each other and also similar to activated carbon

and mixture while activated carbon and mixture were different from each

other. In the week 7 where the inflow concentration was lower all filters

performed similar to each other.

4.6 Pathogen and indicator organisms residence time

Over a period of 56 days filters were fed with mixture of artificial greywater

spiked with high concentration of model pathogen and indicator organisms. In

order to examine the pathogen and indicator organisms resident time, filters

were fed for a week with only artificial greywater.

4.6.1 Salmonella spp.

No Salmonella spp. was found in the effluent in the final week. The average

Salmonella spp inflow concentration in week 8 was around 7 107 CFU mL

-1.

4.6.2 Enterococcous faecalis

The average E. faecalis inflow concentration from week 8 was around 105 and

the concentration of E. faecalis decreased a lot in all filters. The biochar

effluent concentration was around 10 CFU mL-1

, except the second day which

was 22 CFU mL-1

.The effluent from bark contained high concentration of

E. faecalis in the first day, 342 CFUmL-1

, but decreased to 20 in second day

and 10 for the last three days. In activated carbon filters the effluent

concentration was around 10 CFU mL-1

in the first day but increased to 100 in

the second day and then decreased to 43, 20 and 18 in the following days. In

mixture filters the concentration in the first day was around 10 CFU mL-1

and

decreased to half by time. In general the concentration of E. faecalis was

38

around 4 log10 lower than the average inflow concentration in week 8

(Figure 22).

Figure 22. Average Log10 concentration of E. faecalis. in biochar, bark,

activated carbon and mixture over the last six days of experience.

4.6.3 Bacteriophage фX174

The average ɸX174 inflow concentration in week 8 was around

2 106

PFU mL-1

. In effluent from biochar, the concentration of ɸX174 was

around 3000 PFU mL-1

in the beginning of week 9 and decreased by time to

1500 PFU mL-1

. Concentration of ɸX174 in bark effluent was around

12000 PFU mL-1

in the beginning and decreased to 9000 PFU mL-1

in the last

day. Activated carbon filters showed the lowest concentration, around

300 PFU mL-1

in the first day, but it increased to around 1000 PFU mL-1

for

few days and 400 PFU mL-1

on the last day. In mixture filters the concentration

was changing between 5000 and 10000 PFU mL-1

. The concentration of

ɸX174 was generally between 2 and 3 log10 lower than average inflow

concentration in week 8 (Figure 23).

0

1

2

3

4

5

6

0 1 2 3 4 5 6

Avera

ge log

10 c

oncentr

atio

n

(CF

U m

L-1

)

Time (days)

BC B

AC M

Av.log10 inf.conc. W8 detection limit

39

Figure 23. Average Log10 concentration of ɸX174 in biochar, bark,

activated carbon and mixture over the last six days of experience.

4.6.4 Bacteriophage MS2

The average MS2 inflow concentration in week 8 was around

3.5 107

PFU mL-1

. Concentration of MS2 in effluent from biochar and

activated carbon was around 105 in the first day and it did not decrease by time.

In bark and mixture filters the concentration was around 106 all the time except

for the second day on which no MS2 was found in the effluent. In general the

concentration of MS2 was around 2 log10 lower than the average inflow

concentration in week 8 (Figure 24).

0

1

2

3

4

5

6

7

0 1 2 3 4 5 6

Avera

ge lo

g10 c

oncentr

atio

n

(PF

U m

L-1

)

Time (days)

BC B

AC M

Av.log10 inf.conc. W8 detection limit

40

Figure 24. Average Log10 concentration of MS2 in biochar, bark,

activated carbon and mixture over the last six days of experience.

0

1

2

3

4

5

6

7

8

0 1 2 3 4 5 6

Avera

ge lo

g10 c

oncentr

atio

n

(PF

U m

L-1

)

Time (days) BC B

AC M

Av.log10 inf.conc. W8 detection limit

41

5 Discussion

5.1 Organic loading of greywater

The organic load can affect the reduction of pathogens and indicator

organisms. Adsorption of bacteria can decrease at high organic load due to the

occupation of adsorption sites by organic matter (Lalander et al., 2012).

In this study, the organic load of artificial greywater was 76 g BOD5 m-2

day-1

(240 g COD m-2

day-1

or 5 g L-1

). This is much higher than the organic load in

Dalahmeh et al. (2012) study which was 14 g BOD5 m-2

day-1

(890 mg L-1

),

corresponding to what has been reported for greywater in Palestine, Jordan and

Israel (Dalahmeh et al., 2012). The reduction of thermotolarant fecal coliforms

in Dalahmeh et al. (2012) was higher in comparison to this study. Lens et al.

(1994) study demonstrates similar results in removal of fecal coliforms and

fecal streptococci at 312-410 mg COD L-1

. The results of Lalander et al. (2012)

study revealed that reduction of E. coli at higher organic loads reduces in

biochar and sand filters but increases in bark filters.

5.2 Filter material

Physical characteristics of filter material determine its microbial removal

efficiency (Dalahmeh et al., 2012) and the physical characteristic of filter

material is based on the starting organic material and also the process of

carbonization or pyrolysis in case of biochar and activated carbon

(Downie et al., 2009). Biochar and bark were originated from biomass (salix

and pine) and activated carbon from mineral (black coal).

42

In comparison to sand, the most common filter material, filter materials

examined in this study had lower bulk density and thus higher specific surface

area which leads to higher removal efficiency. The particle density and thus

bulk density, specific surface area and porosity of biochar were lower than bark

and activated carbon. The water content was also very low compared to bark

and activated carbon and mixture. In mixture of bark and activated carbon,

bulk density was lower than the other filters, water content was high but

porosity was unrealistically high (Table 1). The surface charge of the filter

material particles can affect the adsorption of pathogens and indicator

organisms. Biochar and bark are negatively charged material and activated

carbon is positively charged material.

The effective particles size in small values provides bigger specific surface

area which can lead to biofilm formation and thus higher reduction of bacteria

(Dalahmeh et al., 2012). The lowest effective particles size observed in

activated carbon (1.2 mm) and the highest in mixture (1.5 mm) while it was the

same in biochar and bark (1.4 mm) (Table 1). Dalahmeh et al. (2012) and

Lalander et al. (2012) demonstrated that bark filters efficiently remove the

pathogens and indicator organisms contrariwise to this study. Bark filters

studied by Lewis et al. (1995), also showed very good result in removal of

MS2 and E. faecalis. Weak performance of bark filters studied in this project

can be a result of physical parameter such as porosity, strain capacity and

retention time. Lewis et al. (1995) show that the bark filters with more moist

and compact material are considerably more efficient than the less compact

one.

5.3 Residence time and tracer recovery

The tracer recovery and residence time show the characteristics of filter

material in term of adsorption and degradation of organic matter, pathogens

and indicator organisms. In longer residence time pathogens and indicators are

in contact with the filter material for longer time, which can lead to bacterial

biofilm formation and more bacterial reduction. The tracer recovery shows the

adsorption capacity of the filter material, the lower the recovery the higher is

the binding capacity (Dalahmeh et al., 2012). Tracer recovery is related to the

specific surface area and surface activity of the filter media.

Biochar, bark and activated carbon achieved 100% recovery while mixture

showed 25% of tracer recovery. It can be explained by the lower effective size

and higher specific surface area of the filter material. Bark and biochar have

43

the same effective size and also the same mean residence time, activated

carbon has the lower effective size among all the biofilters, while mixture has

the highest effective size. In the study done by Dalahmeh et al. (2012) biochar

and B bark showed 50% tracer recovery and only sand reached the 95% NaCl

recovery. Biochar and bark were similarly weak in tracer recovery and it was

explained by big surface area of biochar and the active surface of bark.

Activated carbon had the highest mean residence time, twice as longer as in

biochar and bark. It was 90 hours for both biochar and bark at hydraulic load of

0.032 m3 m

-2 day

-1. Lens et al. (1994) reported the same results (92 h) for

0.5-m tall bark filter at different Hydraulic load (0.01 m3 m

-2 day

-1). Dalahmeh

et al. (2012) reported 43 and 16 hours for 0.6-m tall bark and biochar filters in

the same hydraulic load as in this study. Dalahmeh et al. (2012) and (Lens et

al., 1994) demonstrated higher reduction of pathogens and indicator organims.

5.4 Pathogen and indicator organisms reduction

The mechanisms of bacterial removal is combination of biofilm formation,

straining and adsorption in the filters (Dalahmeh et al., 2012, Stevik et al.,

1999) while virus removal maintains by adsorption to the filter material similar

to small particles (Lalander et al., 2012). A biofilm is a population of

microorganisms adhered to a surface (Stepanović et al., 2004) in response to

environmental signals such as nutritional condition and consist of initiation,

maturation, maintenance and dissolution stages (O'Toole et al., 2000). Biofilm

continue to develop as long as nutrients are available and then start to detach

(O'Toole et al., 2000) but the detachment is not a well-known stage and the

effect of nutrient shortage have not been investigated. Attachment of

microorganisms to the media surface and formation of biofilm is driven by

various parameters.

Porous media characteristics

Porous media with smaller particle size can lead to higher straining and also

can offer more adsorption sites, which leads to higher pathogenic reduction

(Stevik et al., 2004). Difference between the media charge and bacterial charge

is another parameter to consider. However this electrostatic and van der waals

forces are week and function at short distances (between the bacteria and

media) and at high retention times. Presence of macropores decrease the

hydraulic retention time which leads to lower adsorption (Stevik et al., 2004).

44

Shape of the particles can affect the attachment of bacteria to the surface, for

instance Stevik et al. (2004) mentioned that small platy shape particles have

higher adsorption capacity comparing to other particles.

Bacterial biofilm

Biofilm itself can increase the adsorption of bacteria by offering more

adsorption sites in the biofilm (Stevik et al., 2004).

Bacterial cell surface characteristics

Presence of surface appendages e.g. flagella, fimbriae and pilli leads to

bacterial motility, which enhances the chance of bacteria to meet a possible

adhesion surface. The surface charge of the bacteria is influenced by size and

pH and can also affect the adsorption onto the filter media. Number of

electrostatic charge sites increase by increasing the size and may lead to higher

level of adsorption. Bacterial surface of most species has negative net charge

potential at pH around 7, which is higher than their isoelectric point (ISP). ISP

is the pH at which surface charge changes. ISP for most bacterial species is

around 1.5-4. At pH values higher than ISP, surface charge becomes more

negative whereas at pH values lower than ISP surface charge becomes positive

(Stevik et al., 2004). However, effect of ISP and electrostatic charges in

adsorption becomes more significant at size smaller than 60 nm (фX174 and

MS2 but not Salmonella spp. and E. faecalis) (Scot E. Dowd, 1998) .

Gram staining can influence the adsorption. Polysacharids present on the cell

wall of gram positive bacteria can form hydrogen bonds and dipole-dipole

interactions with media surface, which are stronger than van der Waals forces

between gram negative bacteria and surface media.

Size and shape of the microorganism

Bigger surface area offers more interaction sites which leads to higher

adsorption. Moreover, distance between microorganism and media decrease

when the size of microorganism is bigger than the media particle size.

Stevik et al. (2004) have illustrated better removal of long rod-shaped cells

present in wastewater through porous media.

45

5.4.1 Bacterial reduction

Although both Salmonella spp. and E. faecalis are negatively charged and

facultative anaerobic bacteria that form biofilm on the surfaces, reduction of

Salmonella spp. was slightly better than E. faecalis.

Salmonella spp. is long rode-shaped and motile bacteria (fimbriae and pilli)

which have higher chance to meet available adsorption sites whereas

E. faecalis is cocci shaped and non-motile. On the other hand E. faecalis can

tolerance very extreme conditions, such as starvation but Salmonella spp.

prefer inert medium reach nutrient media for attachment (Stepanović et al.,

2004). Moreover, E. faecalis produce extracellular antimicrobial peptides,

bacteriocin, which results in efficient use of energy and inhibiting competitive

bacteria (Fisher and Phillips, 2009).

In Dalahmeh et al. (2012) study log10 reduction of fecal coliforms was around

1.3 in biochar, similar to this study, while the log10 reduction in bark was 2.4,

higher than this study

Salmonella spp. reduction

Reduction of Salmonella spp. in biochar filters was slightly better than other

filters, 2 log10 reduction or higher, due to the presence of smaller particles

(weight share of 1-1.4 mm particles= 50%).

Activated carbon was expected to perform much better than the other filters in

term of bacterial reduction due to its highest specific area and positively

charged particles. In fact, within the first week 7 log10 reduction of

Salmonella spp. was observed in activated carbon, but from the third week it

decreased to 1 log10 reduction. This is not likely to be the result of big pores

since the rapid flow was not observed during the experiment. The high organic

load engaging the adsorption sites is a more likely cause. Appearance of

anaerobic conditions is another parameter to be considered. Filter material

pores were not constantly filled with water, since the hydraulic load in this

experiment was not enough to saturate the filters completely and filters were

drained by gravity. Aerobic conditions in the filters prepare appropriate

environment for biodegradation and nitrification, which release energy for

microbial activity. Anaerobic condition leads to denitrification, which slows

down the biodegradation process (Inglett et al., 2005). Salmonella spp. is

facultative anaerobic organism which is capable to switch from aerobic

46

respiration to fermentation to derive energy, which means that can tolerate

possible anaerobic condition.

Mixture filters, with lowest specific surface area and lower level of small

particles (weight share of 1-1.4 mm particles= 50%) performed 1 log10

reduction within the first 30 days, similar to the bark, and 3log10 reduction in

the following days similar to activated carbon.

Bark showed the least effect on Salmonella spp. removal. Although the

effective size of bark was the same as biochar and porosity was similar to

activated carbon, it seems that longer retention time is required in bark filters

in order to efficiently remove pathogens.

E. faecalis reduction

Although E. faecalis are big size bacteria, their removal was similar to removal

of bacteriophages.

Bark and mixture performed very weak reduction, 1 log10 reduction or less,

which can refer to the fact that retention time was not sufficient.

In biochar and activated carbon the reduction was 4 log10 in the beginning but

decreased to 1-2 log10, which can be the result of adsorption sites occupation

over the time.

Apparently, E. faecalis biofilm formation was weak in bark filters and it was

more likely removed by adsorbing similar to bacteriophages. E. faecalis

re-growth in filters can be another considerable reason in lower reduction of

E. faecalis (Dalahmeh et al., 2012). Although Entrococcus spp. concentration

in the inflow was below the detection limit in Dalahmeh et al. (2012), it was

detected in the effluent from biochar filters in the concentration around

100 CFU mL-1

due to bacterial re- growth in filters similar to this study. No

Entrococcus spp. was found in effluent from bark filters, while performance of

bark filters was weak in this study.

5.4.2 Bacteriophage reduction

Results showed lower reduction of bacteriophages in comparison to bacteria

due to their small size, фX174: 27 nm and MS2: 27 nm (Collins et al., 2006). It

means that in competition with organic particles and bacteria, bacteriophages

have smaller chance to meet the adsorption sites.

47

Electrostatic interactions between the virus particles and the media strongly

control the fate of bacteriophages in the porous media (Lalander et al., 2012).

Number of electrostatic charges on the particles decrease when the size is

reduced, lowering the affinity between particles and filter media. Moreover,

ISP influence the attachment of small microorganisms (<60 nm) to the surface

more than bigger ones (Pelleïeux et al., 2012). Bacteriophages behave as

negatively charged particles at pH values higher than ISP, but become

positively charged particles at pH values lower than ISP.

Bacteriphage фX174 ISP (6.6-6.8) is higher than MS2 ISP (3.5) (Langlet et al.,

2008, Collins et al., 2006). It means that MS2 have more negative charge sites

on the surface in all the filters (activated carbon> biochar> mixture> bark)

while фX174 have less negative charge sites in three filters

(activated carbon> biochar> mixture> bark). In bark filters pH was lower than

фX174 ISP, which resulted in a change in surface charge from negative to

positive, which resulted in low adsorption. Biochar, activated carbon and

mixture had much higher pH values than the ISP of the bacteriophages studied

in this project. In bark filters pH was close to ISP of фX174 but higher than

ISP of the MS2. Activated carbon has positive charge and can absorb

negatively charged particles. Ever since the bacteriophages could not meet

their ISP they continued behaving as negatively charged particles and could

adsorb better to activated carbon comparing to biochar and bark which have

negatively charged surface.

Bacteriophage фX174 reduction

Reduction of фX174 was low in bark and mixture filters, 1 log10 reduction or

less which can be an effect of low retention time and big distance to adsorption

sites (Collins et al., 2006). Although the pH values in bark filters were lower

than фX174 ISP which, it could not overcome the retention time limitation.

Activated carbon showed 6 log10 reduction in the first week, which decreased

to 4 log10 reduction in following three weeks and 2 log10 reduction during the

remaining experimental period. Activated carbon had positively charged

particles, the highest specific surface area and more adsorption free sites,

which resulted in high reduction in the beginning, but after the occupation of

free adsorption sites the reduction decreased over time.

48

Biochar also performed higher reduction in the first week, 4 log10 reduction,

which decreased to 2 log10 reduction in the second week and to

1 log10 reduction in the rest of experience.

Bacteriophage MS2

Low reduction of MS2 agree with other research (e.g Collins et al. (2006)

Pelleïeux et al. (2012)). Although the reduction of MS2 was lower than

фX174, filters followed the pattern similar to фX174.

Bark and mixture showed 2 log10 reduction and biochar and activated carbon

showed 3 log10 reduction in the first week, which decreased by time to less than

1 log10 reduction in the last week.

5.5 Log reduction and inflow concentration

Statistical results showed that the concentration of the studied organisms in the

inflow significantly affects the reduction of Salmonella spp., E. faecalis and

ɸX174 in all filters, with one exception: reduction of E. faecalis in biochar

filters. Higher concentration of Salmonella spp. and ɸX174 in the inflow led to

higher reduction. Similar results were found by Bengtsson and Lindqvist

(1995), Stevik et al. (2004) and Fletcher (1977). One possible reason is that

higher concentration of bacteria and bacteriophage can increase the collisions

between bacteria/bacteriophage and media surface result in higher adsorption.

Furthermore, biofilm expanding increase the number of adsorption sites.

The reduction of E. faecalis decreased under higher inflow concentration. One

possible reason is that biofilm formation was not the main removal mechanism

in E. faecalis and the reduction was similar to particle adsorption on the media

surface. On the other hand, E. faecalis enter the stationary phase (non-growth

phase) and decrease the cell size under extreme condition such as lack of

nutrients due to high concentration of bacteria (Portenier et al., 2003), which

also could have decreased the adsorption rate.

5.6 Pathogen and indicator organisms residence time

Filters were fed with artificial greywater without any addition of pathogen and

indicator organisms for the period of five days to examine the presence and

survival of pathogens and indicator organisms in the filters.

49

No Salmonella spp. was detected in the effluent of any of the biofilters, which

illustrates that biofilm did not reach the detachment stage within the

experimental time. Nonetheless, the pathway of bacterial detachment from the

surface is not known and requires further investigation.

Concentration of E. faecalis was very low in effluent form all filters; especially

bark filters mainly due to the trapping than microbial or chemical processes.

Bacteriophages found in higher concentration than bacteria and among the

bacteriophages, MS2 detected in higher concentration, especially in bark

filters. At current hydraulic load bacteriophages appeared to detach from the

surface, most likely due to their weak bonds with the media.

5.7 Reuse possibilities of filtered water

Activated carbon filters could provide the requirements for restricted irrigation

while effluent from other biofilters can be used only for subsurface irrigation.

Activated carbon filters could reach the requirements for unrestricted irrigation

considering bacteria and ɸX174 removal. However, since the reduction of MS2

was lower than standards for root crops and low growing crops, it can be

applied for leaf crops and high growing crops.

Considering the installation and maintenance, aeration system should be

provided to avoid anaerobic conditions. Furthermore, the results of feeding the

columns with tap water illustrated that under rainy conditions filters would not

release considerable concentration of bacteria. However, concentration of

ɸX174 can increase close to the limit values and concentration of MS2 could

exceed the limits. Preparing protection wall around the system can help to

avoid the release of pathogens to the water bodies.

50

6 Conclusion This study contributes with further knowledge of the aerobic treatment’s

efficiency in removal of pathogen and indicator organisms from greywater.

The study demonstrated that physical and chemical characteristics of the filters

affect the reduction of studied microorganisms. It was found that biochar and

activated carbon filters were slightly better than bark and mixture filters in

removing the studied microorganisms. However, the characteristics of the

pathogen and indicator organisms appeared to influence the reduction to a

greater extends than the filter media properties. Little to no reduction of

bacteriophage MS2 (0-0.5 log10) was observed in all four filter media, while

Salmonella spp., E. faecalis and ɸX174 were reduced by 1-2 log10. It was also

observed that the reduction of Salmonella spp. and ɸX174 correlated to the

inflow concentration in all filter media.

It was also found that viruses were released from the filters for several days

after the last feeding, while bacteria were not. There is thus a risk that viruses

are released from this type of filters long after a virus related outbreak.

Although the treatment proved to reduce the pathogen and indicator organisms

to certain extend at the given hydraulic and organic loading rate, the intended

use of the treated greywater is the key factor when considering the minimal

reduction required in the treatment.

The four filter materials studied would provide water suitable for subsurface

irrigation, while only water treated in activated carbon filters would meet the

WHO requirements for restricted irrigation.

51

References

AL-JAYYOUSI, O. R. 2003. Greywater reuse: Towards sustainable water

management. Desalination, 156, 181-192.

ANWAR, A. H. M. F. 2011. effect of greywater on soil characteristics.

BENGTSSON, G. & LINDQVIST, R. 1995. Transport of Soil Bacteria Controlled

by Density-Dependent Sorption Kinetics. Water Resources Research, 31,

1247-1256.

BERGER, C. 2012. Biochar and activated carbon filters for greywater treatment –

comparison of organic matter and nutrients removal. Swedish University

og Agricultural Science.

BERGEY, D. H. & HOLT, J. G. 1994. Bergey's Manual of Determinative

Bacteriology, USA, Williams & Wilkins.

CHEN, H. M., ZHENG, C. R., TU, C. & SHEN, Z. G. 2000. Chemical methods

and phytoremediation of soil contaminated with heavy metals.

Chemosphere, 41, 229-234.

COETZEE, M. A. A., P.2, R.-V. M. M. & BADENHORST, J. 2010. The Effect of

Hydraulic Loading Rates on Nitrogen Removal by Using a Biological

Filter Proposed for Ventilated Improved Pit Latrines.

COLLINS, K. E., CRONIN, A. A., RUEEDI, J., PEDLEY, S., JOYCE, E.,

HUMBLE, P. J. & TELLAM, J. H. 2006. Fate and transport of

bacteriophage in UK aquifers as surrogates for pathogenic viruses.

Engineering Geology, 85, 33-38.

DALAHMEH, S. S., PELL, M., VINNERÅS, B., HYLANDER, L. D., ÖBORN, I.

& JÖNSSON, H. 2012. Efficiency of Bark, Activated Charcoal, Foam and

Sand Filters in Reducing Pollutants from Greywater. Water, Air, & Soil

Pollution, 223, 3657-3671.

DIXON, A. M., BUTLER, D. & FEWKES, A. 1999. Guidelines for Greywater Re-

Use: Health Issues. Water and Environment Journal, 13, 322-326.

DOWNIE, A., KROSKY, A. & MUNROE, P. 2009. Physical Properties of

Biochar. In: J.LEHMANN & S.JOSEPH (eds.) Biochar for

52

Environmental Management: Science and Technology. London:

Earthscan.

EPA 2002. Onsite Wastewater Treatment Systems Manual.

ERIKSSON, E., AUFFARTH, K., HENZE, M. & LEDIN, A. 2002. Characteristics

of grey wastewater. Urban Water, 4, 85-104.

ERIKSSON, E. & DONNER, E. 2009. Metals in greywater: Sources, presence and

removal efficiencies. Desalination, 248, 271-278.

EU 2009. The rapid alert system for food and feed (RASFF)- Annual Report.

European commission.

FAO 2007. Coping with Water Scarcity.

FAO 2010. The wealth of waste.

FAOWATER 2008. Water at a glance.

FISHER, K. & PHILLIPS, C. 2009. The ecology, epidemiology and virulence of

Enterococcus. Microbiology, 155, 1749-57.

FLETCHER, M. 1977. The effects of culture concentration and age, time, and

temperature on bacterial attachment to polystyrene. Canadian Journal of

Microbiology, 23, 1-6.

FRIEDLER, E., KOVALIO, R. & BEN-ZVI, A. 2006. Comparative study of the

microbial quality of greywater treated by three on-site treatment systems.

Environ Technol, 27, 653-63.

GÜVEN KAYAOGLU & ØRSTAVIK, D. 2004. Virulence Factors of

Enterococcus faecalis: Relationship to Endodontic Disease.

sagepublications.

HENZE, M. 2008. Biological Wastewater Treatment: Principles, Modelling and

Design, IWA.

INGLETT, P., REDDY, R. & CORSTANJE, R. 2005. Anaerobic Soils. In:

HILLEL, D. (ed.) Encyclopedia of Soils in the Environment. Florida,

USA: Elsevier.

JEPPESEN, B. 1996. Domestic greywater re-use: australia's challenge for the

future. Desalination, 106, 311-315.

LALANDER, C., DALAHMEH, S., JÖNSSON, H. & VINNERÅS, B. 2012.

Hygienic quality of artificial greywater subjected to aerobic treatment - a

comparison of three filter media at increasing organic loading rates.

LANGLET, J., GABORIAUD, F., DUVAL, J. F. L. & GANTZER, C. 2008.

Aggregation and surface properties of F-specific RNA phages:

Implication for membrane filtration processes. Water Research, 42, 2769-

2777.

LENS, P. N., VOCHTEN, P. M., SPELEERS, L. & VERSTRAETE, W. H. 1994.

Direct treatment of domestic wastewater by percolation over peat, bark

and woodchips. Water Research, 28, 17-26.

LEWIS, G. D., LOMAX, T. D. & KIMBERLEY, M. 1995. Removal of virus

particles, bacteria and bovine serum albumin from water by steam-

exploded Pinus radiata bark. Water Research, 29, 1689-1693.

53

LI, Z., BOYLE, F. & REYNOLDS, A. 2010. Rainwater harvesting and greywater

treatment systems for domestic application in Ireland. Desalination, 260,

1-8.

MOREL, A. & DIENER, S. 2006. Greywater management in low and middle-

income countries: review of different

treatment systems for households or neighbourhoods. Swiss Federal Institute of

Aquatic Science and

Technology Eawag.

MURRAY, P. R., PFALLER, M. A., ROSENTHAL, K. S., MURRAY, P. R.,

ROSENTHAL, K. S. & PFALLER, M. A. 2002. Bacterial structure

[Online]. Missouri. Available:

http://micro.digitalproteus.com/morphology2.php.

NEGAHBAN-AZAR, M., SHARVELLE, S. E., STROMBERGER, M. E.,

OLSON, C. & ROESNER, L. A. 2012. Fate of Graywater Constituents

After Long-Term Application for Landscape Irrigation. Water Air and

Soil Pollution, 223, 4733-4749.

O'TOOLE, G., KAPLAN, H. B. & KOLTER, R. 2000. Biofilm formation as

microbial development. Annual Review of Microbiology, 54, 49-79.

O'TOOLE, J., SINCLAIR, M., MALAWARAARACHCHI, M., HAMILTON, A.,

BARKER, S. F. & LEDER, K. 2012. Microbial quality assessment of

household greywater. Water Research, 46, 4301-4313.

OTTOSON, J. 2004. Comparative analysis of pathogen occurrence in wastewater-

management strategies for barrier function and microbial control. KTH.

OTTOSON, J. & STENSTRÖM, T. A. 2003. Faecal contamination of greywater

and associated microbial risks. Water Research, 37, 645-655.

PELLEÏEUX, S., BERTRAND, I., SKALI-LAMI, S., MATHIEU, L.,

FRANCIUS, G. & GANTZER, C. 2012. Accumulation of MS2, GA, and

Qβ phages on high density polyethylene (HDPE) and drinking water

biofilms under flow/non-flow conditions. Water Research, 46, 6574-6584.

PINTO, U., MAHESHWARI, B. L. & GREWAL, H. S. 2010. Effects of greywater

irrigation on plant growth, water use and soil properties. Resources,

Conservation and Recycling, 54, 429-435.

PORTENIER, I., WALTIMO, T. M. T. & HAAPASALO, M. 2003. Enterococcus

faecalis– the root canal survivor and ‘star’ in post-treatment disease.

Endodontic Topics, 6, 135-159.

R.ELEY, A. 1996. Microbial food poisoning, UK, Chapman & Hall.

RHEN, M. 20007. Salmonella: Molecular Biology and Pathogenesis, UK.

RIEMANN, H. P. A. & CLIVER, D. O. 2006. Foodborne Infections and

Intoxications, Elsevier Academic press.

ROSE, J. B., SUN, G.-S., GERBA, C. P. & SINCLAIR, N. A. 1991. Microbial

quality and persistence of enteric pathogens in graywater from various

household sources. Water Research, 25, 37-42.

SAHLSTROM, L., ASPAN, A., BAGGE, E., DANIELSSON-THAM, M. L. &

ALBIHN, A. 2004. Bacterial pathogen incidences in sludge from Swedish

sewage treatment plants. Water Research, 38, 1989-1994.

54

SANTOS, C., TAVEIRA-PINTO, F., CHENG, C. Y. & LEITE, D. 2012.

Development of an experimental system for greywater reuse.

Desalination, 285, 301-305.

SCOT E. DOWD, S. D. P., SOOKYUN WANG AND M. YAVUZ 1998.

Delineating the specific influence of virus isoelectric point and size on

virus adsorption and transport through sandy soils. Applied and

environmental microbiology.

SELLNER, M. 2009. Combined greywater treatment using a membrane bioreactor.

STEFANAKIS, A. I. & TSIHRINTZIS, V. A. 2012. Effects of loading, resting

period, temperature, porous media, vegetation and aeration on

performance of pilot-scale vertical flow constructed wetlands. Chemical

Engineering Journal, 181-182, 416-430.

STEPANOVIĆ, S., ĆIRKOVIĆ, I., RANIN, L. & S✓VABIĆ-VLAHOVIĆ, M.

2004. Biofilm formation by Salmonella spp. and Listeria monocytogenes

on plastic surface. Letters in Applied Microbiology, 38, 428-432.

STEVIK, T. K., AA, K., AUSLAND, G. & HANSSEN, J. F. 2004. Retention and

removal of pathogenic bacteria in wastewater percolating through porous

media: a review. Water Res, 38, 1355-67.

STEVIK, T. K., AUSLAND, G., HANSSEN, J. F. & JENSSEN, P. D. 1999. The

influence of physical and chemical factors on the transport of E. coli

through biological filters for wastewater purification. Water Research, 33,

3701-3706.

UNU-INWEH. 2011-2012. The global water crisis: addressing an urgent security

issue [Online]. Canada: United Nations University –Environment and

Health (UNU-INWEH). Available:

http://www.inweh.unu.edu/WaterSecurity/documents/WaterSecurity_FIN

AL_Aug2012.pdf.

WHO 2006a. Guidelines for the safe use of wastewater, excreta and greywater.

policy and regulatory aspects.

WHO 2006b. WHO guidlines for the safe use of wastewater, excreta and

greywater. wastewater use in agriculture.

VOHLA, C., ALAS, R., NURK, K., BAATZ, S. & MANDER, U. 2007. Dynamics

of phosphorus, nitrogen and carbon removal in a horizontal subsurface

flow constructed wetland. Sci Total Environ, 380, 66-74.

ZAPATER, M., GROSS, A. & SOARES, M. I. M. 2011. Capacity of an on-site

recirculating vertical flow constructed wetland to withstand disturbances

and highly variable influent quality. Ecological Engineering, 37, 1572-

1577.

SLU Institutionen för energi och teknik Box 7032 750 07 UPPSALA Tel. 018-67 10 00 www.slu.se/energyandtechnology

SLU Department of Energy and Technology Box 7032 S-750 07 UPPSALA SWEDEN Phone +46 18 671000